Welcome back! I’m Gary, and on my YouTube channel, I create AI-generated cat videos with adorable stories, vibrant visuals, and always a surprising twist. These fun, family-friendly shorts consistently hit around 90% retention—a rare metric to achieve on 60-second videos.
If you’ve tried making YouTube Shorts, you know keeping a viewer engaged for a full minute is no small feat. But I’ve cracked the code, and in this post, I’ll walk you through exactly how to make these videos step by step—including how to make them 100% monetizable on YouTube.
🎯 Why Monetization Matters for AI Content Creators
YouTube has recently become more selective with monetizing AI-generated content. Videos featuring excessive gore, horror, or overly shocking content are often demonetized or shadowbanned.
But my videos are fully monetized, because they follow platform guidelines: safe, engaging, and original content. If you focus on clean, story-driven videos like I do, your chances of monetization go way up.
🧠 Step 1: Brainstorming the Idea & Scriptwriting
Every viral video starts with a strong story. For my Disneyland-themed cat video, I came up with a cute plot: a mother cat takes her child to an amusement park—but there’s a twist!
To create your own story:
- Start with a simple plot idea.
- Use ChatGPT to write a full script using this prompt:
textCopyEdit“Write a short story with 18 lines of natural, easy-to-understand dialogue. Keep the plot under 100 words, include a surprising ending, and make it suitable for a family audience.”
This balance of brevity, structure, and clarity helps ChatGPT generate scripts perfect for YouTube Shorts
🐱 Step 2: Designing the Cat Characters (Using Leonardo AI)
I asked ChatGPT to create visual prompts for:
- A mother cat with white fur and sky-blue eyes
- A child cat in pastel rainbow pajamas with cloud patterns
I used Leonardo AI to generate the images using those prompts.
🔧 Pro Tip:
If you don’t like the outfit or style, ask ChatGPT: “Give me 5 more cute outfit prompts for a mother cat in a family-friendly animation.”
Keep refining until you find a look you love.
🖼️ Step 3: Generating Scene Images with Consistent Characters
Now that the characters are designed, I use ChatGPT to generate prompts for each scene of the story, ensuring character consistency.
Use this format:
cssCopyEditGenerate a 3D render of the white-furred mother cat with sky-blue eyes wearing [outfit], standing next to the child cat wearing pastel rainbow pajamas. They are at [scene setting]. Both characters are smiling.
👉 This ensures the same visual identity across all scenes.
Use Leonardo AI again to generate images from each prompt. Select the best images and regenerate if needed.
🧏♀️ Step 4: Creating Voiceovers with 11 Labs
Voiceovers bring your characters to life. I use 11 Labs, which has the most realistic AI voices.
Here’s how I do it:
- Paste each character’s dialogue line-by-line.
- For the mother, I use the “Kelly” voice.
- For the child, I use the “XII” voice.
📌 Pro Tip: Batch your dialogue and generate all voiceovers together. If any line sounds off, regenerate until it matches your tone.
✂️ Step 5: Editing the Video in CapCut
Here’s where the magic happens. I use CapCut to put everything together into a polished, high-retention video.
Key Editing Steps:
- Import all images and voiceovers.
- Sync them chronologically according to your script.
- Add a slow zoom to each image to create subtle motion.
- Use transitions like “White Flash” (for scene changes) and “Wave Right” (for surprises).
- Add captions using the Auto Caption tool—then customize font and color.
- Apply auto color grading to enhance vibrancy.
- Insert sound effects (giggles, crowd sounds, amusement park ambiance).
- Add cheerful background music from upbeat.com.
Your audio structure should look like this:
- Track 1: Voiceovers
- Track 2: Sound effects
- Track 3: Additional effects
- Track 4: Background music
With this structure, your video stays clear, engaging, and professionally edited.
🎬 Bonus: Tips for Character Scenes with Two Cats
Scenes with two characters are trickier. You need perfect visual and emotional balance.
Here’s what to include in your prompt:
- Character appearances (from previous image prompts)
- Their posture (e.g., “child cat looking up at mother cat”)
- Their expression (e.g., “both cats laughing with joy”)
- Background setting (e.g., “inside a colorful amusement park”)
You can ask ChatGPT: “Write a scene description where a mother and child cat are holding hands at Disneyland. Keep the design consistent with earlier scenes.”
Use Leonardo AI again and regenerate until it looks perfect.
💰 How These Videos Stay Monetizable
Remember: YouTube wants content that’s family-friendly, engaging, and safe. My videos have:
- No gore or violence
- No shock-value tricks
- Original content and characters
- High retention and engagement
This means YouTube not only monetizes them, but they also perform better in the algorithm.
📈 Results: 90% Retention & Growing Views
When I checked the analytics for the Disneyland cat video, it had:
- 90%+ retention rate
- A high click-through rate (CTR)
- Strong engagement in the comments section
That’s the formula for viral growth!
🎁 Wrapping It Up
You now know the full workflow:
- Write a story using ChatGPT
- Design characters with Leonardo AI
- Generate scene visuals
- Record voices using 11 Labs
- Edit like a pro in CapCut
- Maintain monetization with clean content
With these steps, you can create your own viral cat AI Shorts—or adapt the formula for other genres too!
💬 Have questions or want my exact prompts? Leave a comment or DM me on social.
👉 Don’t forget to subscribe for more AI video tutorials!
Leave a Reply